Cluster the stores
Stores grouped by role (destination, convenience, outlet) and by category affinity. Each cluster gets its own assortment shape.
The AI groups your stores, watches your SKU mix, and flags the laggards and the gaps — with the margin and stock-turn trade-offs attached. Your category leads approve; the live view updates.
One weekly category review cycle replaces the quarterly "full reset" that nobody finishes.
Stores grouped by role (destination, convenience, outlet) and by category affinity. Each cluster gets its own assortment shape.
ABC × XYZ per cluster. Lifecycle stage (launch, core, tail, markdown). Contribution to margin, turn, and customer acquisition.
Add, remove, deepen, narrow. Each proposal comes with the expected margin delta, the turn delta, and the risk of stock-out during the change.
Category lead approves the plan. The agent tracks execution across stores, flags deviations, and reports outcomes against the proposal.
Auto-clustering by sales profile, catchment, footfall, category performance — with manual override per store.
Contribution and variability per SKU, per cluster. Updated continuously, not once a quarter.
Launch, core, tail, markdown. Each stage has its own rules for depth, facings, and pricing.
Where the assortment is thin vs. demand. Where two SKUs compete for the same shelf role.
SKU deletions proposed with the margin and turn trade-off. Not by gut — by numbers you can check.
The plan is a first-class object. Every move is tracked, flagged if delayed, reported against the original expectation.
Assortment reads from the forecast and pricing, writes back to the till and the supplier portal.
Talk to the founders. Get early access, honest pricing, and a direct line to the team.